Completed

Roland Berger Campus Challenge 2025

Roland Berger 2025.06Team Lead
Strategy
Market Analysis
Humanoid Robot
Go-to-Market
Unit Economics

Strategy Case · Humanoid Robotics Go-to-Market

Project Overview

Strategy consulting case on the humanoid robotics industry. We mapped the market, sized TAM/SAM/SOM, pressure-tested early beachheads, and designed a staged go-to-market plan with unit-economics checkpoints (BOM, service model, utilization). The work connected tech readiness with commercial viability, and translated insights into a crisp storyline for executives.

What I Did

  • Structured market sizing (top-down & bottom-up): task inventory → serviceable tasks → adoption ramp by segment.
  • Built unit-economics model: BOM ranges, service contracts, utilization, maintenance, and payback sensitivity.
  • Customer discovery & segmentation: manufacturing, logistics, security/inspection, eldercare; scored by feasibility & ROI.
  • Defined go-to-market sequencing: pilot → lighthouse wins → vertical playbooks → partner ecosystem.
  • Partnership thesis: integrators, sensor/actuation suppliers, cloud/AI stacks, safety/insurance stakeholders.
  • Risk & regulation scan: workplace safety, human-in-the-loop requirements, data/telemetry governance.
  • Crafted executive deck: MECE storyline, exhibit design, and ‘so-what’ recommendations.

Reflection

Humanoid robotics is less a single “product” and more a stack: hardware affordability & reliability (actuation, power density, hands/dexterity), perception & control across messy, long-tail tasks, and an operations layer (fleet mgmt, remote assist, safety/guardrails) that actually makes deployments viable. Near-term beachheads are “dull-dirty-dangerous” and constrained environments: pallet handling, inspection rounds, night-shift security, basic material movement. Success metrics aren’t just ‘can it walk’ but: task success rate (TSR), mean-time-to-intervention (MTTI), time-to-first-task-library (TTFTL), and safety events per 1k hrs. Economically, capex must be amortized via service models (RaaS) with predictable uptime; value creation comes from labor substitution on low-variance tasks plus safety/quality benefits. The credible path I see is ‘humanoid as a platform’ with curated task libraries and human-in-the-loop tele-assist, not pure autonomy day-one. Integration matters: APIs to WMS/MES/ERP, site-mapping, and operator training drive adoption more than model benchmarks. The lesson for me as a strategy candidate: sanity-check hype with unit economics; pick beachheads where environmental control and measurable ROI exist; and design the partnership fabric early so scaling is not an afterthought.